Blank, Rebecca M. "The Effects of Double-Blind versus Single-Blind Reviewing: Experimental

 

Blank, Rebecca M. "The Effects of Double-Blind versus Single-Blind Reviewing: Experimental
Evidence from the American Economic Review," American Economic Review, 81 (December
1991), pp. 1041-1967.

The summary statment leading this article is: "The results from a randomized experiment
conducted at The American Economic Review on the effects of double-blind versus single-blind
peer reviewing on acceptance rates and referee ratings indicate that acceptance rates are lower
and referees are more critical when the reviewer is unaware of the author's identify. These
patterns are no significantly different between female and male authors. Authors at top-ranked
universities and at colleges and low-ranked universities are largely unaffected by the different
reviewing practices, but authors at near-top-ranked universities and at nonacademic institutions
have lower acceptance rates under double-blind reviewing. (JEL A12, C93)" (p. 1041)

The focus is clear: do reviewers respond to the authors of papers or to the substance of papers?
Do Big Names at prestige universities have an edge? Are males favored over females in getting
articles into prestige journals? In discussing the problems and her review of the literature, Blank
cites Ceci and Peters work (1980; 1982) who suggested that peer review was a random procedure
and subject to substantial institutional bias. Unfortunately, the data on these important points are
meagre.

The experiment conducted here: Blank was concerned with the acceptance rates of women in
AER and she designed this experiment. Blank's proposal was accepted in May, 1987 and
continued through May 1989. Half the papers received by the editors in that period were
reviewed in a single blind system and half were reviewed using a double blind system.

The conclusions: "1) There is a different pattern of acceptance rates under a single-blind
reviewing system than under a double-blind system. The most striking effect is that acceptance
rates are lower and referee rteports are more critical under a double-blind system.

"2) Thrre are differences in the acceptance rates across institutional categories between
single-blind and double-blind reviewing. In general, authors at the top-ranked institutions and
those ar college or low-ranked institutions are little affected. However, authors at institutions
ranked (high) have lower acceptance rates under blind reviewing, as do foreign authors and
authors from nonacademic institutions.

"While there is some indication in these data that women do slightly better underr a double-blind
system, both in terms of acceptance rates and referee ratings, these effects are relatively small
and statistically insignificant. Thus, this paper provides little evidence that moving to a
doublt-blind reviewing system will substantially increase the acceptance rate for papers by
female economists."(pp. 1063-1064)

It should be noted that this experiment was NOTY kept secret. Referees were aware of what was
going on and this knowledge alone might have influenced their attempts at fairness.